Predictive posteriors under hidden confounding

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Meixide, Carlos García, Insua, David Ríos
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911203385671680
author Meixide, Carlos García
Insua, David Ríos
author_facet Meixide, Carlos García
Insua, David Ríos
contents Predicting outcomes in external domains is challenging due to hidden confounders that potentially influence both predictors and outcomes. Well-established methods frequently rely on stringent assumptions, explicit knowledge about the distribution shift across domains, or bias-inducing regularization schemes to enhance generalization. While recent developments in point prediction under hidden confounding attempt to mitigate these shortcomings, they generally do not provide principled uncertainty quantification. We introduce a Bayesian framework that yields well-calibrated predictive distributions across external domains, supports valid model inference, and achieves posterior contraction rates that improve as the number of observed datasets increases. Simulations and a medical application highlight the remarkable empirical coverage of our approach, nearly unchanged when transitioning from low- to moderate-dimensional settings.
format Preprint
id arxiv_https___arxiv_org_abs_2507_05170
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Predictive posteriors under hidden confounding
Meixide, Carlos García
Insua, David Ríos
Methodology
Predicting outcomes in external domains is challenging due to hidden confounders that potentially influence both predictors and outcomes. Well-established methods frequently rely on stringent assumptions, explicit knowledge about the distribution shift across domains, or bias-inducing regularization schemes to enhance generalization. While recent developments in point prediction under hidden confounding attempt to mitigate these shortcomings, they generally do not provide principled uncertainty quantification. We introduce a Bayesian framework that yields well-calibrated predictive distributions across external domains, supports valid model inference, and achieves posterior contraction rates that improve as the number of observed datasets increases. Simulations and a medical application highlight the remarkable empirical coverage of our approach, nearly unchanged when transitioning from low- to moderate-dimensional settings.
title Predictive posteriors under hidden confounding
topic Methodology
url https://arxiv.org/abs/2507.05170